docs: Update documentation with new functionality

Author: nicholas-cecere

README.md

@@ -18,12 +18,14 @@ This Terraform provider allows you to manage LiteLLM resources through Infrastru
 
 ## Using the Provider
 
+To use the LiteLLM provider in your Terraform configuration, you need to declare it in the `terraform` block:
+
 ```hcl
 terraform {
   required_providers {
     litellm = {
-      source  = "local/litellm/litellm"
-      version = "1.0.0"
+      source  = "litellm/litellm"
+      version = "~> 1.0.0"
     }
   }
 }
@@ -32,34 +34,32 @@ provider "litellm" {
   api_base = var.litellm_api_base
   api_key  = var.litellm_api_key
 }
+```
+
+Then, you can use the provider to manage LiteLLM resources. Here's an example of creating a model configuration:
 
-# Example model configuration for AWS Bedrock
-resource "litellm_model" "claude_aws_bedrock" {
-  model_name            = "claude-3.5-sonnet-v2"
-  custom_llm_provider   = "bedrock"
-  base_model            = "anthropic.claude-3-5-sonnet-20241022-v2:0"
-  tier                  = "paid"
-  mode                  = "completion"  # Specify the model mode
-  aws_access_key_id     = var.aws_access_key_id
-  aws_secret_access_key = var.aws_secret_access_key
-  aws_region_name       = var.aws_region
-
-  input_cost_per_million_tokens  = 4.0
-  output_cost_per_million_tokens = 16.0
+```hcl
+resource "litellm_model" "gpt4" {
+  model_name          = "gpt-4-proxy"
+  custom_llm_provider = "openai"
+  model_api_key       = var.openai_api_key
+  model_api_base      = "https://api.openai.com/v1"
+  base_model          = "gpt-4"
+  tier                = "paid"
+  mode                = "completion"
+  
+  input_cost_per_million_tokens  = 30.0
+  output_cost_per_million_tokens = 60.0
 }
 ```
 
-### Model Mode
-
-The `mode` attribute in the `litellm_model` resource allows you to specify the intended use of the model. Available options are:
+For full details on the `litellm_model` resource, see the [model resource documentation](docs/resources/model.md).
 
-- `completion`
-- `embeddings`
-- `image_generation`
-- `moderation`
-- `audio_transcription`
+### Available Resources
 
-This attribute is optional. If not specified, it will not be included in the API request.
+- `litellm_model`: Manage model configurations. [Documentation](docs/resources/model.md)
+- `litellm_team`: Manage teams. [Documentation](docs/resources/team.md)
+- `litellm_team_member`: Manage team members. [Documentation](docs/resources/team_member.md)
 
 ## Development
 

docs/resources/model.md

@@ -11,14 +11,20 @@ resource "litellm_model" "gpt4" {
   model_api_key       = "sk-your-api-key"
   model_api_base      = "https://api.openai.com/v1"
   api_version         = "2023-05-15"
-  base_model         = "gpt-4"
-  tier               = "paid"
-  tpm                = 100000
-  rpm                = 1000
+  base_model          = "gpt-4"
+  tier                = "paid"
+  mode                = "completion"
+  tpm                 = 100000
+  rpm                 = 1000
   
   # Cost configuration (per million tokens)
   input_cost_per_million_tokens  = 30.0    # $0.03 per 1k tokens = $30 per million
   output_cost_per_million_tokens = 60.0    # $0.06 per 1k tokens = $60 per million
+
+  # AWS-specific configuration (if applicable)
+  aws_access_key_id     = "your-aws-access-key-id"
+  aws_secret_access_key = "your-aws-secret-access-key"
+  aws_region_name       = "us-west-2"
 }
 ```
 
@@ -28,61 +34,40 @@ The following arguments are supported:
 
 * `model_name` - (Required) The name of the model configuration. This will be used to identify the model in API calls.
 
-* `custom_llm_provider` - (Required) The LLM provider for this model. Examples include:
-  * `openai`
-  * `anthropic`
-  * `azure`
-  * `cohere`
+* `custom_llm_provider` - (Required) The LLM provider for this model (e.g., "openai", "anthropic", "azure", "bedrock").
 
-* `model_api_key` - (Required) The API key for the underlying model provider.
+* `model_api_key` - (Optional) The API key for the underlying model provider.
 
-* `model_api_base` - (Required) The base URL for the model provider's API.
+* `model_api_base` - (Optional) The base URL for the model provider's API.
 
 * `api_version` - (Optional) The API version to use for the model provider.
 
 * `base_model` - (Required) The actual model identifier from the provider (e.g., "gpt-4", "claude-2").
 
-* `tier` - (Optional) The usage tier for this model. Valid values are:
-  * `free`
-  * `paid`
-  Default is `free`.
+* `tier` - (Optional) The usage tier for this model. Valid values are "free" or "paid". Default is "free".
+
+* `mode` - (Optional) The intended use of the model. Valid values are:
+  * `completion`
+  * `embeddings`
+  * `image_generation`
+  * `moderation`
+  * `audio_transcription`
 
 * `tpm` - (Optional) Tokens per minute limit for this model.
 
 * `rpm` - (Optional) Requests per minute limit for this model.
 
-* `input_cost_per_million_tokens` - (Optional) Cost per million input tokens. This will be automatically converted to the per-token cost required by the API. For example:
-  * Set to `30.0` for a cost of $0.03 per 1k tokens ($30 per million)
-  * Set to `3.0` for a cost of $0.003 per 1k tokens ($3 per million)
-
-* `output_cost_per_million_tokens` - (Optional) Cost per million output tokens. This will be automatically converted to the per-token cost required by the API. For example:
-  * Set to `60.0` for a cost of $0.06 per 1k tokens ($60 per million)
-  * Set to `6.0` for a cost of $0.006 per 1k tokens ($6 per million)
-
-## Additional Configuration Options
+* `input_cost_per_million_tokens` - (Optional) Cost per million input tokens. This will be automatically converted to the per-token cost required by the API.
 
-The following optional parameters are also supported:
+* `output_cost_per_million_tokens` - (Optional) Cost per million output tokens. This will be automatically converted to the per-token cost required by the API.
 
-* `timeout` - (Optional) Request timeout in seconds.
-* `stream_timeout` - (Optional) Streaming request timeout in seconds.
-* `max_retries` - (Optional) Maximum number of retry attempts.
-* `organization` - (Optional) Organization identifier for the model provider.
-* `region_name` - (Optional) Region name for region-specific providers.
+### AWS-specific Configuration
 
-### Provider-Specific Options
+* `aws_access_key_id` - (Optional) AWS access key ID for AWS-based models.
 
-#### Google Vertex AI
-* `vertex_project` - Project ID for Google Vertex AI.
-* `vertex_location` - Location for Google Vertex AI resources.
-* `vertex_credentials` - Credentials for Google Vertex AI authentication.
+* `aws_secret_access_key` - (Optional) AWS secret access key for AWS-based models.
 
-#### AWS
-* `aws_access_key_id` - AWS access key ID.
-* `aws_secret_access_key` - AWS secret access key.
-* `aws_region_name` - AWS region name.
-
-#### IBM WatsonX
-* `watsonx_region_name` - Region name for WatsonX services.
+* `aws_region_name` - (Optional) AWS region name for AWS-based models.
 
 ## Attribute Reference
 
@@ -92,7 +77,10 @@ In addition to the arguments above, the following attributes are exported:
 
 ## Import
 
-Model configurations can be imported using the model name:
+Model configurations can be imported using the model ID:
 
 ```shell
-terraform import litellm_model.gpt4 gpt-4-proxy
+terraform import litellm_model.gpt4 <model-id>
+```
+
+Note: The model ID is generated when the model is created and is different from the `model_name`.